NILU: TR 11/2003 REFERENCE: U-103002
DATE: NOVEMBER 2003 ISBN: 82-425-1518-2 ISSN: 0807-7185
C o n t e n t s
Editorial . . . 5
Nomenclature . . . 5
New developments . . . 5
Zardoz hardware . . . 5
Directory structure . . . 5
VINTERSOL directory tree . . . 5
Daily updated plots . . . 6
Near-Real-Time processing of ozone sonde data for ECMWF . . . 7
Current and recent projects . . . 8
The VINTERSOL campaign . . . 8
Overview of the VINTERSOL projects . . . 8
The GMES-GATO project . . . 9
The ASSET project . . . 9
The QUILT project. . . 9
The GOA project . . . 9
The UFTIR project. . . 10
The MAPSCORE project. . . 10
The CANDIDOZ project and the ERA-40 data set . . . 10
The EUPLEX project . . . 10
The QUOBI project . . . 10
The PVC project and the ERA-15 data set . . . 10
The COSE project . . . 10
The COZUV and AerOzClim projects. . . 11
Network for the Detection of Stratospheric Change (NDSC) . . . 11
The ENVISAT Cal-Val Database . . . 12
Introduction . . . 12
Web access . . . 12
Data formats. . . 12
System description . . . 12
Plots . . . 13
The NILU team . . . 14
Existing data and software . . . 14
THESEO and THESEO 2000 Experimental Data . . . 14
SESAME Experimental Data. . . 14
EASOE experimental data . . . 14
Satellite data. . . 14
ECMWF data on 2.5 ˚
×2.5 ˚ grid. . . 15
Trajectories. . . 15
PV at stations . . . 15
ECMWF data on 1.125 ˚
×1.125 ˚ grid . . . 15
Extraction of data on 1.125 ˚
×1.125 ˚ grid . . . 16
Graphical presentation of data . . . 18
Plotting of isentropic data on 2.5 ˚
×2.5 ˚ grid . . . 20
Plotting of isobaric data at 2.5 ˚
×2.5 ˚ resolution . . . 21
Plotting of trajectories . . . 22
Plotting of T
106data . . . 23
Plotting of TOMS data . . . 24
Plotting of TOVS data . . . 25
Plotting of ozonesonde data . . . 26
Time series of data on 2.5 ˚
×2.5 ˚ grid . . . 27
Time series of T
106data . . . 28
Time series of PV at individual stations. . . 29
Listing of data for single days . . . 29
How to make XY diagrams . . . 30
Other programs . . . 31
A summary of all the programs . . . 31
On-line services . . . 32
NADIR on the Web . . . 33
Who to contact at NILU? . . . 33
Access to ECMWF data. . . 33
Acknowledgements . . . 33
Editorial
his issue of NADIR NEWS is meant to give information on new developments and products at NADIR since the previous issue (December 1998) and in conjunction with the VINTERSOL campaign. Most of the information giv- en in the previous issue has been kept in this issue so that it constitutes an independent source of information. For those who are not familiar with the NADIR data centre it can be mentioned that NADIR was put into operation in con- junction with the European Arctic Stratospheric Ozone Experiment in 1991-92. NADIR is an abbreviation for NILU’s Atmospheric Database for Interactive Retrieval. NADIR now has approx. 500 users from all over Europe as well as USA, Canada, Japan and New Zealand. There are more than 300 login sessions per day during the most intensive winter peri- ods. A first version of Nadir News 2003 was issued in early spring in connection with the startup of the ASSET project.
The current version contains a few updates and corrections, and is prepared for printing.
Nomenclature
We use fonts of different colours and styles in order to improve the readability of the text. In the body text we use this font when we refer to directory names, file names and text lines in text files which are to be taken literally. This font is used for placeholders of text or file names. Such text should not be typed literally but replaced with text that applies to your case. Pro-
grams and scripts are referred to by this font. Commands that you type are shown in this font. If a command contains placeholders, they are given in this font. Links to web pages and ftp sites are given in red colour, like this:
http://www.nilu.no
The PDF version of this document have hyperlinks to all re- ferred web pages, so you can access them just by clicking on the link.
New developments
ince the previous issue of NADIR NEWS (December 1998) a number of new developments have taken place at NADIR. The most important new developement is the ENVISAT Cal-Val database. The following sections describe these new items in more detail. The information given in previous issues is also included, so this issue gives a com- plete description of the NADIR services.
Zardoz hardware
zardoz.nilu.no is a Sun clone with 2 Sparc II processors running at 400MHz. The IP address is 128.39.104.1. It runs the Solaris 7 operating system. zardoz is equipped with approx. 600 GB of disk space.
Directory structure
A new directory structure has been implemented at NADIR.
This has been done in order to make navigation easier for our customers and also to make the maintenance and back-up of the file systems easier. Most project related directories have been moved to directories located under /nadir/projects. Projects that are not directly linked to the coordinated European cam-
paigns are found under /nadir/projects/other. Here one finds projects such as APE, GODIVA, Leewave, OSDOC, SAONAS, GOA and TOPOZ as well as several others.
For the ENVISAT Cal-Val and the ASSET databases, datafiles
are located under /nadir/esa/data and
/nadir/eu/asset/data respectively.
VINTERSOL directory tree
A directory tree has been set up for the VINTERSOL cam- paign. The VINTERSOL data will be stored under /nadir/projects/vintersol.
Ozonesonde data, which are collected mainly as a part of the Match campaign organised through the QUOBI project, will be stored in /nadir/projects/vintersol/data/o3sondes.
T
S
Under the main VINTERSOL directory, data from the related project will be stored in subdirectories such as
/nadir/projects/vintersol/euplex and
/nadir/projects/vintersol/quilt. Coordinators for those projects that have not yet arranged with directories should con- tact the NADIR team as soon as possible.
Daily updated plots
On zardoz you can find ready-made plots of a number of para- meters that are updated on a daily basis. The plots are stored as GIF files. These products are found in various subdirectories under the directory /nadir/plots.
Maps of isobaric ECMWF fields
Analysis and forecast maps of temperature and geopotential height at the 14 standard isobaric levels from 1000 to 5 hPa are made every day. The maps are stored as colour GIF files in /nadir/plots/isobaric_gif/yymmdd. Figure 1 shows an ex- ample of such a map.
Maps of isentropic ECMWF fields
Analysis and forecast maps of temperature and potential vorti- city at isentropic levels are stored as colour GIF files in /nadir/plots/isentrop_gif/yymmdd. Figure 2 shows an ex- ample of a potential vorticity map.
Maps of T
106data
Analysis maps of various parameters at T106 resolution (1.125 × 1.125 degrees are produced on a daily basis. These plots are available under /nadir/plots/t106. There are maps of T and GPH at isobaric levels and of PV and T at isentropic levels. The maps are stored in the GIF format. Figure 3 shows an example of such a map.
ECMWF products on the web
The maps of isobaric, isentropic and T106 data described above,
are also available through the web site http://nadir.nilu.no/ecmwf. In addition to these, it is also possi- ble to create trajectory plots similar to those made by the trajplo program described on page 22.
Through the web site, it is also possible to download preextract- ed T106 data in NASA-Ames format. As a part of the ENVISAT CalVal database project, a routine for extraction of met-data into HDF format was developed. This program is called hdf- mars and is described in detail on page 18. It is possible to ac- cess this program through the web site and thereby extract met- data on demand - solely by the use of a web browser.
The web site is only available for those who have signed the ecmwf4 protocol. Use the same username and password as you do on zardoz.nilu.no to enter the site.
Maps of filaments
In conjunction with the former METRO project, the group at CNRS put into operation a service for daily maps with analyses and forecasts of potential vorticity. Plots of filaments in GIF Figure 1. 120 hours forecast of temperature at 30hPa valid for 12
UT on 13 December 2002.
Figure 2. A 192 hours potential vorticity forecast map valid for 16 December 2002 at the 950K isentropic level (approx. 4-8 hPa).
Figure 3. Map of south polar potential vorticity for 3 October 2002, during the major warming that destroyed a large part of the vortex.
format are produced using a high horizontal resolution MIMOSA advection model (18 × 18 km) of PV developed at Service d’Aéronomie du CNRS1. The model starts from ECMWF PV analysis on 1st November and computes the advection of PV using daily ECMWF analyses and forecasts available on NA- DIR. In order to follow the diabatic evolution of PV during the winter, a relaxation toward the large-scale ECMWF PV field is applied with a 10 days relaxation time. Daily plots of filaments are produced for 7 isentropic levels, 350, 380, 400, 435, 475, 550 and 675 K and for 00, 24, 48, 72, 96 and 120 hours forecast.
Temperature maps at 475K are also produced on the same for- mat for interpretation. Figure 4 shows one example of such a map. These maps are found as GIF files in
http://www.aerov.jussieu.fr/~fgoutail/MIMOSA.html.
Total ozone maps
The WMO real time ozone mapping unit at the University of Thessaloniki produces daily maps of total ozone based on the GO3OS, i.e. the Dobson and Brewer network of spectropho- tometers. These map are found as postscript files in /nadir/plots/wmo. Figure 5 shows such a map.
The Antarctic vortex
During the south polar ozone hole season NILU runs a service for the WMO that calculates certain parameters on the develop- ment of the Antarctic vortex. Among these parameters are min- imum temperatures, vortex area etc. Plots of these parameters are presented on the web. This service has been in operation since 2000 and will continue during the VINTERSOL period.
The plots can be found here:
http://www.nilu.no/projects/nadir/o3hole
Figure 6 shows an example of a plot of the area of the vortex.
Near-Real-Time processing of ozone sonde data for ECMWF
At the joint ECMWF/WMO expert meeting on realtime ex- change of ground based ozone measurements, held at ECMWF in 1996, the requirements for ozone data for NWP (Numerical Weather Prediction) in realtime were outlined. ECMWF has de- veloped the operational data assimilation system to include ozone retrieval from SBUV and GOME, and requires independ- ent high quality profile data from ground-based systems. For daily validation, monitoring and troubleshooting, it was found that ozone sonde measurements should be available at the cen- tre within 12-24 hours after the sounding.
NILU has collected ozone measurements from Arctic balloon flights through the Nadir database since 1988. Files are trans- ferred and stored in the NASA-AMES 2160 format, and an au- tomatic script has been set up to convert incoming data into the CREX format that is used at ECMWF. This script also performs a series of data quality checks, and can do simple corrections on
1. A. Hauchecorne, M. Marchand, S. Godin and C. Souprayen, A high resolution ad- vection model for the interpretation of ozone filaments observed in lower stratospheric ozone lidar profiles at mid-latitudes, Proceedings of the European Workshop on Mes- oscale Processes in the Stratosphere, Bad Tölz, Germany, 9-11 November 1998, pages 105 - 110.
Figure 4. 24 hours forecast map of PV at 550K valid for 7
December 2002. This map has been made with a domain filling trajectory model running at CNRS.
Figure 5. Total ozone for 8 December 2002 based on the GO3OS and GOME.
Figure 6. Area of the south polar vortex at 475K. The red curve shows the development during the unusual 2002 ozone hole season.
erroneous input files.
The CREX-converting routine is a modulised Perl script that automatically runs every 15 minutes on zardoz.nilu.no. On each run:
• The program recursively traverses a given directory.
• New incoming files since last run are identified, and for each file;
• A format check is done with dataex. Does it qualify as a proper NASA- AMES 2160 file?
• If so, the data values are loaded into a Perl object.
• Needed meta-data are retrieved from the header of the NASA-AMES file.
• At each pressure level of the sounding, the values are made sure to be legal and reasonable.
• Checks are made on pressure values, and temperatures inside and out- side of the Styrofoam box.
• Errors are logged, and sometimes causes rejection of the file.
• The total ozone column is approximated
• A CREX file is made with ozone pressure, total pressure, outside tem- perature, relative humidity, time after launch, wind speed and wind direction for each pressure level.
• The final output is stored in a separate directory.
At the end of each run, a log file is updated to keep track of old and new files.
No qualitative changes are made to the sonde data, but the series of pressure levels are made monotonic if needed and bad data lines are skipped in the output. The latter is normally caused by negative pressure or negative time-after-launch values.
ECMWF downloads new CREX files from zardoz every 30 minutes. All ozone soundings that are uploaded to NADIR are therefore converted and transferred to the weather centre in Reading within a maximum of 45 minutes.
The script is currently used for ozone sonde stations that submit data through the VINTERSOL project. This near-real-time product is only available to ECMWF at the moment, but NILU is interested in contributing and developing this resource through future projects.
Current and recent projects
In this section we present some of the current stratospheric ozone projects that are part of the VINTERSOL campaign and other projects that are currently using the NADIR database. Some recent, but completed projects are also presented.
The VINTERSOL campaign
VINTERSOL (Validation of INTERnational Satellites and study of Ozone Loss) is a major European field campaign stud- ying stratospheric ozone. VINTERSOL ( "Winter sun" in the Scandinavian languages) will take place from late 2002 until mid 2004. It is the latest major European field campaign to study ozone loss. There has been three previous European cam- paigns: the European Arctic Stratospheric Ozone Experiment (EASOE); the Second European Stratospheric Arctic and Mid- latitude Experiment (SESAME); and the Third European Strat- ospheric Experiment on Ozone (THESEO). Like them, VIN- TERSOL relies jointly on support from national funding agencies and from the Environment and Sustainable Develop- ment programme of EC DG Research. An important dimension for VINTERSOL is the involvement of several new European satellite instruments. Measurements from the ERS-2 GOME satellite instrument (operational since 1995) and from the POAM III instrument on the SPOT IV satellite (operational since 1998) will continue to be used. However, in addition, measurements from the ODIN satellite (launched in February 2001) and ESA’s new ENVISAT satellite (launched in March 2002) will be validated and, in time, analysed. VINTERSOL is thus being mounted in conjunction with the validation cam- paign for ENVISAT satellite, and it will significantly extend the scope and duration of the validation activities, so enhancing the quality of the measurements made by these satellite instru-
ments. The increasing international dimension to earth observa- tion studies is also evident, as there will be cooperation with the validation campaigns for the NASA SAGE III (SOLVE-2) and the NASDA ILAS-2 satellite instruments. There are three main phases to VINTERSOL in which detailed studies of atmospher- ic processes will be made:
• intensive Arctic ozone loss studies in the 2002/03 winter/spring;
• ozone loss studies in the Antarctic winter and spring 2003; and
• balloon and aircraft studies in the tropics in early 2004.
More information on VINTERSOL and a planning document can be found on the web site of the European Ozone Research Coordinating Unit, http://www.ozone-sec.ch.cam.ac.uk/.
Overview of the VINTERSOL projects
The VINTERSOL campaign consists of a number of projects funded by the European Commission (DG Research) and by various national agencies. The table below lists these projects.
Projects participating in or linked to VINTERSOL
Acronym Coordinator URL
APE-Infra Leopoldo Stefanutti, IROE/CNR
http://ape.iroe.fi.cnr.it/APE- Infra.htm
CIPA Niels Larsen, DMI http://web.dmi.dk/pub/CIPA/
More information on these projects can be found on the web pages of the Ozone Secretariat;
http://www.ozone-sec.ch.cam.ac.uk/
The GMES-GATO project
Global Monitoring for Environment and Security - Global At- mospheric Observations (GMES-GATO) is an EU funded con- serted action, aiming to develope a strategy for global observations of atmospheric composition within GMES. Issues to be adressed include greenhouse gases, stratospheric ozone and urban pollution. The scope of the project is therefore to cover all aspects of atmospheric change on global, regional and local scale. Databases is one out of five topics in GMES-GATO, and issues like user needs and harmonisation between different databases will be addressed. The project is therefore expected to make a significant impact on the future developements of the NADIR data centre. GMES-GATO started 1st of February 2003 and will run for 24 months. More information can be found at- the official web site: http://www.nilu.no/gmes-gato
Information of the cooperation between the European Union
and the European Space Agency through GMES can be found at the official web site: http://www.gmes.info
The ASSET project
ASSET (Assimilation of ENVISAT Data) is a European (EU) initiative to exploit and develop earth observation data from En- visat using data assimilation. The project has two major activi- ties and is aiming to exploit the satellite data in Numerical Weather Prediction (NWP) and investigate the distribution of chemical species. A third supporting activity covers data man- agement, and the NADIR datacentre will be used to make these value-added data available to the scientific community. The project runs from January 1st 2003 until the end of 2005 and aims to (a) develop a European capability for chemical and UV forecasting and (b) provide analyses for coupled climate/chem- istry studies. More information on the ASSET project can be found at the official web site: http://darc.nerc.ac.uk/asset/
The ASSET database builds on the system developed for the ENVISAT CalVal database, and the assimilated products are therefore accessible through a searchable web interface. The ASSET dataset will initially only be available to the project partners, but the idea is to make it public after completion. The ASSET database is found at: http://nadir.nilu.no/asset
The QUILT project
QUILT (Quantification and Interpretation of Long-Term UV- Visible Observations of the Stratosphere) started up in January 2001 and will last through December 2003. The EU project ad- dresses stratospheric ozone depletion in particular, but is also focused on better exploitation of existing data, and develop- ment of new long-term observing capacites. QUILT uses the ex- isting ground-based, satellite and balloon borne UV-visible data as well as 3D atmospheric modelling tools for quantifying ozone loss in the past, to monitor its development in the present and to investigate its relation to active halogen and nitrogen species.
The official QUILT web page is found at http://nadir.nilu.no/quilt and gives a general introduction to con- cepts, aims and achieved goals. Password protected pages are also available to project partners, and allows the consortium members to upload and download datafiles through the web in- terface.
The GOA project
GOME Assimilated and Validated Ozone and NO2 Fields for Scientific Users and for Model Validation, GOA, was a two year long EU project completed in January 2003. Products such as a 7 year (August 1995- May 2003) dataset of assimilated total ozone, a one year dataset of assimilated and validated ozone profiles and a multi-year dataset of total NO2 were produced by the consortium. A large amount (several thousands) of europe-
EDUCE Gunther Seckmeyer, Univ.
of Hannover
http://www.muk.uni-hanno- ver.de/EDUCE
EUPLEX Fred Stroh, FZ Jülich http://www.nilu.no/euplex HIBISCUS Jean-Pierre Pommereau,
CNRS
http://www.aero.jus- sieu.fr/projet/HIBISCUS MAPSCORE John Remedios,
Univ. of Leicester
http://www.leos.le.ac.uk/map score
POSTA Ulrich Schurath, FZ Karsruhe
http://imk-aida.fzk.de/posta
QUILT Geir Braathen, NILU http://nadir.nilu.no/quilt QUOBI Peter von der Gathen, AWI http://www.nilu.no/quobi TROCCINOX Ulrich Schumann, DLR http://www.pa.op.dlr.de/troc-
cinox
UFTIR Martine de Mazière, IASB http://www.nilu.no/uftir AMIL2DA Thomas von Clarmann, FZ
Karlsruhe
http://www-
imk.fzk.de/imk2/ame/amil2da CANDIDOZ Esko Kyrö, FMI http://fmiarc.fmi.fi/candidoz CARIBIC III Carl Brenninkmeijer, Max
Planch Institute, Mainz
http://www.caribic-atmos- pheric.com
EUROSPICE J. Austin, UK Met Office http://www.metof- fice.com/research/strato- sphere/processes/eurospice GOA Hennie Kelder, KNMI http://www.knmi.nl/goa MOZAIC III Jean-Pierre Cammas,
CNRS
http://www.aero.obs- mip.fr/mozaic
TOPOZ III W. Kouker, DLR http://www-imk.fzk.de/topoz- iii
Projects participating in or linked to VINTERSOL
Acronym Coordinator URL
an ozone sonde data, collected at the NADIR data center, were used to validate the ozone profiles. The GOA datasets will be made available from KNMI in 2003. More information on the project can be found at the official web site:
http://www.knmi.nl/goa
The UFTIR project
The consortium of the EU-project UFTIR is built around the European part of the NDSC FTIR community, complemented with some atmospheric modeling and related laboratory teams.
“Time series of Upper Free Troposphere observations from a European ground-based FTIR network” (UFTIR) started in February 2003 and will last for 30 months. The main objective of the project is to provide trends and time-series of distinct tropospheric and stratospheric target gases such as N2O, CH4, HCFC-22, CO, C2H6 and O3. All time-series will be archived at NADIR. More information on the project is found at:
http://www.nilu.no/uftir
The MAPSCORE project
Mapping of Polar Stratospheric Clouds and Ozone Levels Rel- evant to the Region of Europe (MAPSCORE) is an EU-project focused around ozone depletion over Europe and its relation to the formation of polar stratospheric clouds (PSCs). MAP- SCORE will provide data and tools which will advance the un- derstanding of Polar Stratospheric Cloud (PSC) formation and their control of ozone depletion in the European region. Maps of ozone, related chemical species and PSCs will be delivered to the NADIR database. In this context, NILU is providing the partners with tools for quality assured file-generation and visu- alization, and a password protected web site has been set up to download project files and documents. These pages are found on http://nadir.nilu.no/mapscore, while the official mapscore site is located at: http://www.leos.le.ac.uk/mapscore/
The CANDIDOZ project and the ERA- 40 data set
CANDIDOZ (Chemical and Dynamical Influences on Decadal Ozone Change) is a three year long EU project started up in May 2002. The main objective is to establish a scientific basis for the detection of the earliest signs of ozone recovery due to Montreal protocol and its amendments. This will be achieved by selecting the best long-term ozone and meteorological data sets available (by ECMWF and NCEP). In the context of CAN- DIDOZ, the NADIR data centre is collecting PV and pressure fields on five selected isentropic levels from the ERA-40 data set (ECMWF). A complete time-series from 1957 to 2001 is stored in ASCII format (not NASA-Ames) at zardoz.nilu.no.
Access to these data are so far limited to internal use at NILU, but investigators interested in the data set should contact the NADIR team. More information on the CANDIDOZ project is
found at: http://fmiarc.fmi.fi/candidoz
The EUPLEX project
The “European Polar Stratospheric Cload and Lee Wave Exper- iment” is an EU project started in May 2002 and ending in April 2004. EUPLEX will critically test current hypotheses for the three key processes of Arctic stratospheric ozone depletion chemistry: 1) PSC formation and properties in Lee-wave and synoptic scale PSC, 2) Halogen activation on PSC and 3) Chemical ozone loss. The use of stratospheric airplane meas- urements is central in the project, but the scientific problems are also approached with simulations from various models. The EUPLEX web site is located at: http://www.nilu.no/euplex
The QUOBI project
The EU funded project QUOBI (Quantitative Understanding of Ozone losses by Bipolar Investigations) started up in January 2002 and will last for 36 months. The main objective of the project is to test our quantitative understanding of the chemical mechanisms that destroy ozone in the wintertime Arctic strato- sphere, and to improve the representation of these processes in chemical models of the atmosphere. The experimental basis of this project are two co-ordinated match-campaigns of measure- ments of ozone loss rates, one in the Antarctic polar vortex and one in the Arctic polar vortex. This will be the first time such a campaign takes place in the southern hemisphere. The NADIR data center collectes all ozone-sonde measurements on a near- real time basis, and provides forecasts and analyses of the need- ed meteorological parameters. The QUOBI web pages are lo- cated at: http://www.nilu.no/quobi
The PVC project and the ERA-15 data set
PVC (Polar Vortex Change) was an EU project aiming at stud- ying the climatology of the north Polar vortex. For this project we acquired parts of the ERA-15 reanalysis data set from EC- MWF. This data set dates from January 1979 through February 1994, and we can provide the complete set of daily (noon) tem- perature and wind data for all these years. The horisontal reso- lution is 1.125x1.125 degrees (T106), and the upper level is 10 HPa or 700K. Access to this data set is limited to the PVC part- ners, but investigators interested in this data set should contact the NADIR team.
The COSE project
Compilation of ground based measurements in support of Sat- ellite measurements over Europe (COSE) was an EU project under Theme 3.3 (Centre for Earth Observation) of the Environ-
ment and Climate work programme. This project was the suc- cessor of the previous ESMOS project and could be considered as EU’s contribution to the NDSC network. This project en- compassed many of the ground based measurements carried out in Europe with UV-Vis spectrometers, lidars, FT-IRs and mi- crowave instruments. Further information on the completed project is available at http://www.nilu.no/projects/na- dir/cose/cose.html
The COZUV and AerOzClim projects
COZUV (Coordinated Ozone and UV project) was a joint Nor- wegian project within stratospheric ozone and UV research. It started in January 1999 and ran through 2002. The main objec- tives were to gain increased understanding of the processes that lead to ozone loss in the Arctic and at middle latitudes and to understand how changes in the ozone layer affects the amount of UV radiation that hits the ground. More details can be found at: http://www.nilu.no/projects/cozuv/
AerOzClim (Aerosols, Ozone and Climate) is a natural contin- uation of the work laid down in COZUV. The main objective of AerOzClim is to improve our understanding of aerosol- climate and ozone-climate interactions, by developing and applying global models in combination with analysis of data, to study processes involved, and to provide improved parameterisations for climate models. More information at: http://www.geo- fysikk.uio.no/forskning/atmosfare/prosjekter/AEROZCLIM
Network for the Detection of Stratospheric Change (NDSC)
What is the NDSC?
he Network for the Detection for Stratospheric Change (NDSC) is a set of high-quality remote-sensing re- search stations for observing and understanding the physical and chemical state of the stratosphere and assessing the impact of any stratospheric changes on the underlying trop- osphere and on global climate. The measurement priorities con- cern ozone and key ozone related parameters such as temperature, aerosols and tracers of chemistry and atmospheric motion. The current NDSC network of approx. 50 stations are supported by other existing ground-based monitoring networks, by ozonesondes and by measurements from satellites. Over 100 scientists from 15 countries are involved with NDSC research activities world wide. The NDSC is a major component of the international upper atmosphere research effort and has been en- dorsed by national and international scientific agencies, includ- ing the International Ozone Commission, the United Nations Environment Programme (UNEP), and the World Meteorolog- ical Organization (WMO).
Following five years of planning, instrument design and imple- mentation, the NDSC began network operations in January 1991. For more information on the NDSC one can look up the following web sites:
http://climon.wwb.noaa.gov
http://www.ndsc.ws
http://www.nilu.no/projects/ndsc
Several European scientists participate in the NDSC and the former ESMOS projects, funded by the European Commission, have constituted an essential part of Europe’s contribution to this network. Scientists with an interest in the stratospheric ozone problem are invited to participate in the network. The fol- lowing open letter to the scientific community has been pub- lished by the NDSC Steering Committee:
Open letter to all those interested in acquiring access to public NDSC data
The NDSC Data Protocol states:
Since the nature of detection of small changes requires an ex- tremely high level of measurement confidence, the Data Proto- col recognizes that multiple seasonal analyses may be required for validation of observations from both individual and multiple sites. It is expected that such a procedure shall yield the veri- fiable product referred to as “NDSC data” within a two-year period after acquisition. Co-authorship shall be offered on publications resulting from the verification procedure to those investigators participating in the process. After the above veri- fication period, NDSC data will be available to anyone through centralized scientific data archiving and distribution facilities.
In this spirit, data that has been so verified and given the status of ‘NDSC data’, and is more than two years old, are available to the general public. NDSC datasets are outlined in the NDSC Measurements and Analyses Directory (link to the appropriate web page). Access to this data is through an account on the NDSC database computer. To receive such an account, com- plete the following application:
http://climon.wwb.noaa.gov/www/invite.html
As with any raw data, additional material is usually re- quired to convert raw data into meaningful information.
Therefore it is strongly recommended that data users will con- sult the on-line documentation and reference articles to fully understand the scope and limitations of the instruments and re- sulting data. Scientific users of the data are strongly encouraged to directly contact the NDSC Principal Investigator listed in the data documentation to insure the proper use of specific datasets and ensure the latest and most relevant information relating to the particular data set is being used to help with the interpreta- tion.
T
The ENVISAT Cal-Val Database
NILU operates the database for correlative data that will be used for validation of five of the instruments on ENVISAT, namely AATSR, GOMOS, MERIS, MIPAS and SCIAMACHY. The development of this database started in the spring of 2000, but the work builds on the experience gained through the work on the GOME validation database hosted by NILU since 1995. The database is still undergoing new development, in particular when it comes to graphical tools. The cal-val database represents a new development of NADIR. The access is now through your web browser rather than telnet and ftp. The data format is HDF 4.1r3, rather than the NASA Ames format that we traditionally have used in NADIR.
Introduction
NILU has designed and implemented a system for organizing ground based measurement data, and for retrieval of the same data by scientists that perform comparisons with measurements from the ENVISAT satellite. The work has been performed in close co-operation with ESA and with representatives of the user community. The system is complex since it entails co-op- eration between wide spread scientific communities that have separate and different cultures and methods. In the ESA ENVI- SAT Calibration/Validation effort the measurements of strato- sphere physicists, modellers and mathematicians, marine biologists, and space scientists need to be described within one common frame of reference.
Web access
Access to the database is via this URL:
http://nadir.nilu.no/calval
You need a user name and a password in order to log in. If you need an account, please contact Rita Larsen at NILU
(mailto:[email protected])
Data formats
ESA selected the HDF 4.1r3 file format for the file exchange, based on the established use of this format within ESA and some of the user groups. Main software tools have been devel- oped in FORTRAN, IDL, PHP, SQL and UNIX shell-scripts.
The system uses Red Hat Linux, Apache web-server with PHP server-side scripting, and a MySQL database.
Through extensive co-operation with the ESA project Official Dr. Rob Koopman, the system design has been extended and adapted according to the growing awareness of the user require- ments. The final product has evolved considerably compared to what was envisioned when the Work Statement was created, and may be subject to further evolution.
System description
The system components are here described in a logical order when we follow a data file as it passes from the originator into the storage and forward to an end user. The DS (Data Supplier) needs to sign a data protocol and be registered in the system metadata. This allows the user access to the Cal/Val web site, and gives permission to upload data for one or more projects (AOs).
At the Cal/Val website (http://nadir.nilu.no/calval/secure/) the user will find Metadata Guidelines, file templates and other documents that help with formatting original data into an HDF file. A software tool named ASC2HDF is available for Windows, Linux, Solaris and HPUX users. This tool accepts data and metadata in two simple text files, and will generate an HDF file after extensively testing the input according to predefined rules and values in table.dat. This file contains up-to-date informa- tion on all legal values in each metadata field. Whenever meta- data are updated at the central site, a new version of table.dat is posted on the web.
Short descriptions of the metadata elements are displayed on a set of web pages under the "Browse Metadata" item. These pag- es are linked to the index database, and will always show the lat- est updated legal metadata values.
When the HDF file has been successfully tested at the local site by ASC2HDF, it may be uploaded to the Cal/Val site (/na- dir/esa/incoming) by ftp, or through a web upload page (ht- tp://nadir.nilu.no/calval/secure/upload.php). A set of UNIX shell scripts is started every 5 minutes. These scripts check for new files in the /nadir/esa/incoming directory, and process each file by launching a FORTRAN program named HDF2ASC. The program verifies data and metadata according to the prede- fined rules in the table.dat− the same file that was used by ASC2HDF.
Even files that have been successfully tested by the originator, may be rejected at NILU, mostly due to inconsistencies in the file name (which reflects a subset of the metadata content), or due to duplicate file names or out-of-sequence version num- bers. If the data supplier is not accredited for the AOID listed in the file, the file will also be rejected. An error report will auto- matically be emailed to the data supplier and the owner of the logon name that was used, and the file will be moved to a hidden directory.
If everything checks out correctly, the received HDF file will be moved to a storage file tree starting at /nadir/esa/data, and the file name, upload details and central metadata elements are stored in an index database. The system enforces consistent naming of variables and other metadata elements, and consist- ent spelling of names for people, organisations and sites.
The nadir index database contains the official list of allowed metadata values in the Cal/Val HDF data files, in addition to logs of uploaded/downloaded files, an overview of metadata contents, and the variable list of all accepted HDF data files. All this information is available to dynamic web pages at the Cal/Val web site. The main end user tool on this site is the
“Search Data” page (http://nadir.nilu.no/calval/se- cure/getdata4.php), which allows the user to sort through the data files with advanced criteria selections. Filtering by data supplier, project, location, data source, data type, component and other metadata elements is supported. Data files may also be filtered by a “4-D box algorithm” (any file with data relevant for a given geographical location and time). Furthermore, files can be filtered by submission date and update status.
All data files that match the search criteria are listed in a new web page, with links to HDF data file download, to comments, and to a variable list. In the variable list page the user may select variables and generate an on-line plot. In the file list the user may also select multiple files for download as a tar-ball. The user may save the search criteria in the index database for con- venient re-use at a later time.
Users that have an IDL license may download IDL scripts for HDF data file formatting (excluding the detailed error checking available in the FORTRAN version) and for plotting of data sets from HDF files (http://nadir.nilu.no/calval/secure/idltools.php).
Users that have signed an additional protocol for access to EC- MWF data will find pre-computed T106 ASCII extractions and plots for the last 30 days, as well as plots of isentropic and iso- baric forecasts. There is an on-line tool for extracting T106 data into HDF files. There is also an on-line facility for plotting 10- day back trajectories (based on data calculated at DMI by Dr.
B. Knudsen).
Plots
While logged on to Cal/Val database one can generate plots of datafiles. Below follow two examples of data plots. When you are logged in at the restricted pages you click on the line that says "Search Data". This gives you a table with search criteria that you fill in. In the first example below we have chosen Lo- cation = ALOMAR, Data source type = LIDAR.DIAL and Var- iable name = O3.CONCENTRATION. Clicking on Create file list at the bottom of the page will produce a list of files that satisfy these criteria. In this file list click the column that says Variables for the file you wish to plot. You then get a table of contents for that file and you can select which parameters you wish to plot along the axes. We chose to plot temperature and ozone vs. altitude. Clicking on the Create plot button will give you the plot in Figure 7 visualised on the screen. In the sec- ond example we have chosen PI = Vik and Data Source =
Hy2TH. The search returns a list of files, and we choose to plot the file submitted July 18, 2002. This file contains a 3-dimen- sional field of northward wind-velocity, and we choose to plot this versus longitude and latitude. Clicking on the Create plot button, gives you a plot at 3000K. At the bottom of the page, a drop-down menu appears and you are able to replot the figure at other Theta levels. Figure 8 shows the northward wind- speed at 400K. Please note that this plotting tool is developed for on-line web browsing and to give users a brief overview of what the datafiles contains. The quality of the plots are there- fore not optimised for printing.
Figure 7. Example of an on-line plot created at the ENVISAT Cal/Val database. The figure shows temperature and ozone density as a function of altitude, measured by the DIAL system at Alomar.
Figure 8. Example of an on-line plot created at the ENVISAT Cal/Val database. The figure shows northward wind velocity as a function of longitude and altitude at 400K. The datafile has been made with the program hdf-mars, described on page 18.
The NILU team
The development and operation of the ENVISAT Cal/Val data- base is the result of a joint effort by a large team.
The NILU project leader from the start in the spring of 2000 un- til the spring of 2001 was Dr. Bojan Bojkov. He left NILU April 2001.
The Nguyen Thanh took over as the administrative project manager after Bojan Bojkov.
Dr. Aasmund Fahre Vik has joined the team to assist with the scientific part of the project management and takes part in IDL and FORTRAN programming.
Terje Krognes took part in working out data definitions, data- base structures and metadata guidelines, and has done a major part of the web development in PHP and SQL. He laid out the fundamentals of the system architecture, and interacted exten- sively in the detailed system design. He is also responsible for the user support, and handles updates of metadata in co-opera- tion with the Metadata Board.
Sam-Erik Walker has programmed the main software tools in FORTRAN (ASC2HDF), and supported users with analysis of HDF file formatting problems.
Trygve Bårde has programmed the HDF file processing system in UNIX shell scripts, and assisted with hardware and OS in- stallation.
Roland Paltiel has performed the the hardware and software installation.
Bjørn Gloslie has assisted with database configuration and ad- ministration.
Rita Larsen created the initial design of the Cal/Val web site, and has assisted with communication with the user group in ad- ministrative matters, such as data protocol administration.
Dr. Kjersti Karlsen Tørnkvist has joined the team as an IDL programmer.
Dr. Geir Ole Braathen has contributed as a scientific advisor.
The information given in this chapter is adapted from a techni- cal report (NILU OR 54/2002, ISBN 82-425-1399-6)
Existing data and software
he rest of the document describes data and software that were described in previous issues of NADIR NEWS, but there have been some updates and changes. The section on extraction of T106 met-data has been updated with new developements as part of the ENVISAT CalVal project. The section furthermore contains additional information on efficient use of programs. Datasets of isentropic analyses and forecasts are available all the year and now also for the southern hemisphere. Minor changes and bugfixes are commented where appropriate.
THESEO and THESEO 2000 Experimental Data
The directory tree for the old THESEO data are found under /nadir/projects/theseo. Under this directory there are sub- directories for each of the THESEO related projects, such as Halomax, Metro, O3loss, PSC-analysis, PVC, Stratospheric BrO and Wave. Real time PTU soundings from various stations in the European Arctic and Greenland are to be found in /nadir/projects/theseo/misc/ptu. In addition there are PSC alerts provided by the SAOZ network. Those are to be found in /nadir/projects/theseo/misc/psc_alert.
SESAME Experimental Data
xperimental data from the SESAME campaign can be found in /nadir/sesame/data on zardoz. A detailed description of these data was given in the June 1996 is- sue of NADIR NEWS.
EASOE Experimental data
The experimental data from the EASOE campaign was issued on CD-ROM in 1995. The data are public domain, so the disk can be obtained by contacting the NADIR team at NILU. The data can also be found under /nadir/projects/easoe/data.
Satellite data
TOMS
Data from TOMS (Nimbus-7, Meteor-3, ADEOS and Earth Probe) can be found in subdirectories located under /nadir/satellit/toms. There are data from Nimbus-7 until May 1993 and from Meteor-3 until December 1994. Data from ADEOS go from September 1996 until late June 1997. Data from the Earth Probe started in July 1996 and the instrument is still in operation. Software for plotting of TOMS data is de- scribed in the chapter on plotting software. More than 10000
T
E
files are stored in these directories.
TOVS
TOVS data were provided for the SESAME campaign by CNRM in Toulouse, and these data can be found in /nadir/sesame/data/satellit/tovs. More than 5600 files were submitted through September 1997, when the service was discontinued. Software for plotting is described on page 25.
The data files are compressed, but you can copy the files you need to /nadir/tmp and uncompress them if you want to plot them with the plotting program on zardoz.
ECMWF data on 2.5 ˚
×2.5 ˚ grid
hese are the data that have been available throughout EASOE, SESAME and THESEO and which have also been published on three CD-ROM volumes so far. CD- ROMs are not produced anymore.
Isentropic data
Analyses for 12 UT on the levels 350, 380, 400, 435, 475, 550, 675and 950K are made available by the Danish Meteorological Institute for the geographical area from the North Pole to 30 ˚N.
Pressure and PV are given, and it is possible to calculate tem- perature. This is done by several of the programs developed at NILU, such as the Uniras plotting programs. During the cam- paign phases 1, 2, 3, 4, 5 and 8 days forecasts are available.
These data normally cover the time interval from 1 November to 30 April, but due to the activities around the ENVISAT Cal/val database the isentropic data were available throughout 2002. Analyses and forecasts are found in the directory /na- dir/isentrop/yyyy/mm
where yyyy/mm designate the year and month. In this way we limit the number of files in each directory. Forecasts will be de- leted when they are a few days old in order to limit the number of files.
During the Antarctic test-campaign of the QUOBI project in 2002, isentropic data were also made available for the southern hemisphere. These data will also be produced for the real match campaign in 2003, and they are expected to be available in May.
These files are stored in the same directory as those for the northern hemisphere. Arctic files are named atyymmdd.hhh, and Antarctic files are named asyymmdd.hhh, where hhh is the fore- cast time.
Isobaric data
Analyses for 12 UT on the 14 standard levels 1000, 850, 700, 500,
400, 300, 200, 150, 100, 70, 50, 30,10and 5hPa are made by the Norwegian Meteorological Institute. Available parameters are temperature, geopotential height, zonal wind and meridion- al wind. The geographical area is from the North Pole to 30 ˚N.
These data are collected daily around the year. There are also 1, 2, 3, 4 and 5 days forecasts. The analyses and forecasts are now found in/nadir/isobaric/yyyy/mm.
Trajectories
Ten days backward trajectories arriving at a large number of end points are provided by the Danish Meteorological Institute.
All measurement sites participating in VINTERSOL are in- cluded as well as a grid net of 118 end points covering the area from the North Pole to 30 ˚N. There are data for the same eight levels as for the isentropic fields. To begin with, trajectory data were stored in two ways: a) as large collective files (4.5MBytes) with data for all the end points and b) as a number of small files (approx. 40 kBytes) with data for each end point.
This means that the same data was stored twice. As the amount of data accumulated this started to take considerable disk space.
We now store the trajectory data only as large collective files. If you want to extract data for a single station, you can use a pro- gram called traj, which is described on page 31. The trajecto- ry data are found in /nadir/trajecto/yyyy/mm.
PV at stations
otential vorticity at the seven standard levels is calculat- ed for a number of stations by the Danish Meteorologi- cal Institute by bilinear interpolation in a 1.125˚ × 1.125˚
grid. This gives more exact data than one obtains by extraction from the 2.5˚ × 2.5˚ data. Software for extraction of data for stations of interest is described on page 29. These data are found in /nadir/isentrop/pvatstat/yyyy/mm
As a novelty PV, temperature and geopotential height on the different isentropic levels are stored for each station for both 00 and 12 UT. This makes it easy to extract PV, temperature or geo- potential height for a given level above a station for the whole winter. The data are stored in files named pvyymmdd.tt where tt is the analysis time (00 or 12). The files are not in the NASA Ames format, but they are simple and self-explanatory.
T
P
ECMWF data on 1.125 ˚ × 1.125 ˚ grid
he demand for meteorological data increased during the EASOE and SESAME campaigns, and NILU therefore implemented a routine for daily transfer of global fields at so-called T106 resolution. This corresponds to a maxi- mum latitude/longitude resolution of 1.125˚ × 1.125˚. This routine became operational in early January 1995. We have also acquired data from previous winters (the EASOE winter, the 1992-93 winter and the first SESAME winter) on tape, and these data are also available. The latter data are referred to as archive data in the following. We have developed and implemented programs for extraction of the data and for interpolation of data from model levels to either isobaric or isentropic levels. It is also possible to obtain potential vorticity at isentropic levels.
Daily data
These data originate from ECMWF and are stored in spectral form, so-called T106. These are model analyses that currently contain 60 model levels, with model level 60 being the ground or sea surface. The upper levels are pure pressure levels, and the lower are pure sigma levels, whereas the other levels are a mix- ture of the two. Older data contain less levels, and they do not cover the upper stratosphere. At each level several parameters are stored. At present the available parameters are: temperature (T), zonal wind (U), meridional wind (V), vertical wind (W) and the natural logarithm of the surface pressure (LNSP) (LNSP is only present at level 60). Programs to extract these model level data are available on zardoz, and routines for inter- polation of data to isobaric and isentropic levels are also of- fered. Software routines are described later.
Data are available from 00 UT on 8 January 1995 onwards, with 6 hours intervals, giving data at 00, 06, 12 and 18 hours. Usually the data are received at NILU one or two days after calculation at ECMWF. Some longer delays have been encountered.
A set of interpolating routines have been written to convert data from model levels to more natural units like pressure or theta surfaces. These programs will interpolate to a wide range of surfaces. Available pressure surfaces are: 0.1, 0.14, 0.2, 0.3, 0.5, 0.7, 1, 1.4, 2, 3, 5, 7, 10, 14, 20, 30, 50, 70, 90, 100, 140, 200, 300, 500, 700, 850, 1000, 1013.25 hPa; available theta surfaces are all levels from 200 to 700 K in 5 K steps, from 700 to 1000 K in 25 K steps and from 1000 to 3000 in 250 K steps.
All the parameters contained in the model file can be extracted and interpolated, except for LNSP, which is the logarithm of the surface pressure.
Archive data
hese data also originate from ECMWF and are also stored in spectral from, so-called T106. These are model analyses that contain 31 model levels (19 levels before 1992), with model level 31 (19) being the ground or sea surface.
The upper four levels are pure pressure levels and the lower three levels are pure sigma levels. At each level several param- eters are stored. At the present time the available parameters are: temperature (T), specific humidity (Q), zonal wind (U), meridional wind (V), vertical wind (W) and the natural loga- rithm of the surface pressure (LNSP). Specific humidity is available until 4 April 1995, after when this parameter is only
stored on a Gaussian grid.
Programs to extract these model level data exist on zardoz, and extraction of data to isobaric and isentropic levels is also of- fered (see following chapter).
Data are available from 1 October 1988 to 7 January 1995, after when the daily data take over. Please note that access to EC- MWF data from the various years, are restricted by several pro- tocols.
For converting data from model levels to more natural units like pressure or theta surfaces, a set of interpolating programs have been written. These routines will interpolate to a wide range of surfaces. Available pressure surfaces are the same as for the dai- ly data (see above).
All the parameters contained in the model level data can be ex- tracted and interpolated, except for LNSP, which is the loga- rithm of the surface pressure.
The maximum resolution is 1.125 ˚ × 1.125 ˚ for the T106 data. A lower resolution can be specified to the sp2ll program de- scribed below. This will produce an output file with fewer grid points.
T
T
Extraction of data on 1.125 ˚ × 1.125 ˚ grid
t NILU we have developed programs to extract portions of the T106 data. These programs are based on software provided by ECMWF, but considerable effort has been put into the implementation of these programs and in the development of programs for interpolation of model level data onto isobaric and isentropic levels. We have also developed a code to calculate potential vorticity on isentropic levels. The data are in so-called grib format and in spectral form. In order to get useful data from these files one has to convert to ascii numbers on a latitude longitude grid. We have chosen the NASA Ames format for the ascii files. Thus, a file that you extract from the T106 data will be very similar to the other meteorological data on 2.5˚ x 2.5˚ grid. You can extract either the whole globe or a portion of it.
Daily data at model levels
sp2ll
For extraction of data at model levels, only one program has to be used: sp2ll (spectral to lat./lon.). This is a program that uses the gribex routines provided by ECMWF to “degrib” and ex- tract the T106 spectral coefficients and produce data on a regular latitude and longitude grid. The output from this program is an ascii file in NASA Ames format 3010, a plain 7-bit ASCII file.
The sp2ll program takes the arguments from the command line.
• Syntax
sp2ll yy mm dd hh Var Level Resolution West East North South Outfile
yy mm dd and hh are year, month, day and hour, respectively.
Var is the meteorological parameter you want. Level is the model level number, and it has to be in the range 1 to 31. Res- olution is the geographical resolution in degrees. The best res- olution you can obtain from T106 is 1.125 ˚ × 1.125 ˚, so normally you would give 1.125 here, but you can also specify poorer resolutions, such as 2.5. West is the western limit. It has to be in the range -180 to 180. East is the eastern limit. It has to be in range -180 to 180, and it has to be larger than West. North is the northern limit. It has to be in the range 90 to -90.
South is the southern limit. It has to be in the range 90 to -90, and it has to be smaller than North. Outfile is the name of the file to contain the resulting data.
Invoked without command line parameters, sp2ll prints out a usage list. This is helpful when trying to remember the syntax.
The sp2ll program reads the contents of the environment var- iable MARSPATH. This variable must be set by the user before sp2ll is used. MARSPATH must be set to:
/nadir/t106/yyyy/mm This done with the command:
setenv MARSPATH /nadir/t106/yyyy/mm
This command can be put in your .login file if you plan to ex- tract much model level data.
• Example 1
setenv MARSPATH /nadir/t106/1995/02
sp2ll 95 02 25 18 T 18 1.125 -180 180 90 -90 global.T.dat
This command extracts temperature from 18 UT on 25 Febru- ary 1995 for the whole globe at model level 18. The output is stored in the file global.T.dat.
• Example 2
setenv MARSPATH /nadir/t106/1995/01 sp2ll 95 01 30 06 U 31 1.125 -50 50 90 30 europe.U.dat
This command produces zonal wind from 06 UT on 30 January 1995 for a region stretching from 50˚W to 50˚E and 90˚N to 30˚N, and the result is stored in the file europe.U.dat.
Archive data at model levels
For archive data (i.e. data which has been acquired on tape from ECMWF) the process of extracting data on model levels is iden- tical to the one for daily data.
Daily & Archive data at pressure and theta levels
met-mars
n order to obtain T106 data at pressure and theta levels an additional set of programs have to be used. These are the interpolating programs hy2p, hy2th etc. A normal user will not need to be concerned about the usage of these pro- grams, since a script called met-mars in most cases performs the task of extracting and interpolating by calling the necessary programs. The script met-mars is invoked by the user and will extract and interpolate. It will automatically set MARSPATH, so the user does not need to set this environment variable.
• Syntax
met-mars yy mm dd hh West East North South Resolution Surface Level Variable Outfile
where yy, mm, dd, hh, West, East, North, South and Resolution have the same meaning as above. The latter should be 1.125 (only this resolution works). Surface is the type of surface and can be either th or p. Level is the numer- ical value of the level (see the chapter describing the data on page 15 for allowed levels). Variable is the name of the mete- orological parameter and can be one of the following: T, U, V, W, Z, PV and, in addition for archive data, Q. All of these variables can be interpolated to any of the available surfaces. The out- file from this script is in NASA Ames format number 3010.
A
I
This is a plain 7 bit-ASCII file that can be transferred by ftp or e-mail.
• Example 1
met-mars 90 11 10 12 -180 180 90 30 1.125 th 475 T t901110.12.475
This will extract temperature on a theta surface at 475 Kelvin for the 10 November 12 UT producing a NASA Ames output file named t901110.12.475. The geographical area is from the North Pole to 30˚N.
• Example 2
met-mars 95 8 10 12 -180 180 90 30 1.125 th 435 PV pv950810.12.435
This will give potential vorticity at 435 Kelvin over the North- ern Hemisphere down to 30 ˚N.
• Example 3
Potential Vorticity over the south polar area at a theta surface:
met-mars 95 9 20 18 -180 180 -50 -90 1.125 th 650 PV pv950920.18.650
This gives PV between 50˚S and 90˚S at 650 K.
• Example 4
met-mars 95 9 23 06 -180 180 -60 -90 1.125 p 100 T t950923.06.100
This gives temperature over the south polar area (60˚S to 90˚S) at the 100hPa pressure surface.
• Example 5
met-mars 95 9 26 12 -180 180 90 -90 1.125 th 500 PV pv950926.12.500
This gives a global field of PV at 500 K for 26 September 1995.
At present only the 1.125 degree resolution can be used in the met-mars script, although the sp2ll program can accept any resolution.
The script met-mars is resident in /nadir/bin and can be cop- ied and changed by the experienced user. The script is written in the Bourne shell. This script utilizes the programs sp2ll, hy2pv, hy2th, hy2p, hy2z_th and hy2z_p. 4 versions are available for each of the hy2xx programs since input data con- tain 19, 31, 50 or 60 model levels. Not all kinds of variations are
covered by this script, and some users may have to change the script to fit the individual needs.
Any problems that you might have using these programs should be reported to Aasmund F. Vik at NILU.
HDF-mars
hdf-mars is very similar to met-mars, but the output is returned as an HDF file insted of ASCII. The HDF format is similar to that used in the ENVISAT Cal/Val database system. When us- ing hdf-mars, you are not supposed to indicate any Outfile, since the file-name is automatically generated according to the CalVal metadata definitions. The output from the program is written to the users home directory.
• Syntax
hdf-mars yy mm dd hh West East North South Resolution Surface Level Variable
The arguments are similar to those in met-mars.
Making extractions more time-efficient
Both met-mars and hdf-mars are able to extract all available isentropic or isobaric levels into one file. This is done by giving -1 (minus one) instead of a level. For met-mars, the time con- sumption is almost the same when extracting all levels com- pared to extraction of a single level. For hdf-mars, the time consumption is somewhat larger when all levels are extracted at once, but significantly less than the time needed to extract all levels separatly.
If you are planning to download data on several levels, it is strongly adviced to use the multi-level option. This will surely save you a lot of time.
Pre-extracted T
106data
In connection with the daily updated production of T106 maps described on page 6, a selection of data on 8 isentropic and 20 isobaric levels are premade and stored on zardoz on a daily ba- sis. Data are always available for the last 30 days. The files are located at /nadir/met/extractions/isobaric/yymmdd/ and /nadir/met/extractions/isentropic/yymmdd/. The same files are furthermore available for download through the web page http://nadir.nilu.no/ecmwf
Graphical presentation of data
niras programs have been developed to plot data at T106 resolution. Some of the old Uniras programs have been rewritten, so that they are easier to install, modify and maintain. All programs described in the following can either be run on zardoz, where they have been installed, or you can transfer the source code to your local com- puter and install it there. The syntax of the old programs is as before, but the number of necessary input files has been drastically reduced. In order to plot data at T106 resolution, you first have to extract the data you want to plot, and then start the plotting program.
